2015-01-23  Akim Demaille  <akim@lrde.epita.fr>

	version 3.0.4
	* NEWS: Record release date.

2015-01-23  Akim Demaille  <akim@lrde.epita.fr>

	gnulib: update

2015-01-23  Akim Demaille  <akim@lrde.epita.fr>

	build: re-enable compiler warnings, and fix them
	There are warnings (-Wextra) in generated C++ code:

	  ltlparse.cc: In member function 'ltlyy::parser::symbol_number_type
	  ltlyy::parser::by_state::type_get() const':
	  ltlparse.cc:452:33: warning: enumeral and non-enumeral type in
	  conditional expression
	      return state == empty_state ? empty_symbol : yystos_[state];

	Reported by Alexandre Duret-Lutz.

	It turns out that -Wall and -Wextra were disabled because of a stupid
	typo.

	* configure.ac: Fix the stupid typo.
	* data/lalr1.cc, src/AnnotationList.c, src/InadequacyList.c,
	* src/ielr.c, src/print.c, src/scan-code.l, src/symlist.c,
	* src/symlist.h, src/symtab.c, src/tables.c, tests/actions.at,
	* tests/calc.at, tests/cxx-type.at, tests/glr-regression.at,
	* tests/named-refs.at, tests/torture.at:
	Fix warnings, mostly issues about variables used only with assertions,
	which are disabled with -DNDEBUG.

2015-01-22  Akim Demaille  <akim@lrde.epita.fr>

	tests: c++: fix a C++03 conformance issue
	This fixes test 241 on xLC:

	"input.y", line 42.11: 1540-0274 (S) The name lookup for "report" did not find a declaration.
	"input.y", line 42.11: 1540-1292 (I) Static declarations are not considered for a function call if the function is not qualified.

	where report is:

	  static void
	  report (std::ostream& yyo, int ival, float fval)
	  {
	    yyo << "ival: " << ival << ", fval: " <<  fval;
	  }

	and line 42 is:

	  %printer { report (yyo, $$,       $<fval>$); } <ival>;

	It turns out that indeed this function must not be declared static,
	<http://stackoverflow.com/a/17662745/1353549>.  Let's put it into an
	anonymous namespace.

	Reported by Thomas Jahns.
	http://lists.gnu.org/archive/html/bug-bison/2015-01/msg00059.html

	* tests/actions.at (Qualified $$ in actions): Don't use "static",
	prefer anonymous namespace.

2015-01-20  Akim Demaille  <akim@lrde.epita.fr>

	tests: fix a title
	* tests/conflicts.at: De-overquote.

2015-01-20  Akim Demaille  <akim@lrde.epita.fr>

	c++: reserve 200 slots in the parser's stack
	This is consistent with what is done with yacc.c and glr.c.  Because
	it also avoids that the stack needs to be resized very soon, it should
	help keeping tests about destructors more reliable.

	Indeed, if the stack is created too small, very soon the C++ library
	needs to enlarge it, which means creating a new one, copying the
	elements from the initial one onto it, and then destroy the elements
	of the initial stack: that would be a spurious call to a destructor.

	Reported by Thomas Jahns.
	http://lists.gnu.org/archive/html/bug-bison/2015-01/msg00059.html

	* data/stack.hh (stack::stack): Reserve 200 slots.
	* tests/c++.at: Remove traces of stack expansions.

2015-01-20  Akim Demaille  <akim@lrde.epita.fr>

	tests: be more robust to unrecognized synclines, and try to recognize xlc
	Reported by Thomas Jahns.
	http://lists.gnu.org/archive/html/bug-bison/2015-01/msg00059.html

	* tests/synclines.at (AT_SYNCLINES_COMPILE): Rename as...
	(_AT_SYNCLINES_COMPILE): this.
	Try to recognize xlc locations.
	(AT_SYNCLINES_COMPILE): New.  Skips the test if we can't read the
	synclines.

2015-01-20  Akim Demaille  <akim@lrde.epita.fr>

	tests: fix C++ conformance
	Reported by Thomas Jahns.
	http://lists.gnu.org/archive/html/bug-bison/2015-01/msg00059.html

	* tests/c++.at (Exception safety): Add missing include.
	Don't use const_iterator for erase.

2015-01-18  Akim Demaille  <akim@lrde.epita.fr>

	build: fix some warnings
	Reported by John Horigan.
	http://lists.gnu.org/archive/html/bug-bison/2015-01/msg00034.html

	* src/graphviz.c, src/symtab.h: Address compiler warnings.

2015-01-16  Akim Demaille  <akim@lrde.epita.fr>

	build: avoid infinite recursions on include_next
	On MacOS X 10.5 PPC with Apple's GCC 4.0.1:

	  % uname -a
	  Darwin aria.cielonegro.org 9.8.0 Darwin Kernel Version 9.8.0: Wed Jul 15 16:57:0
	  1 PDT 2009; root:xnu-1228.15.4~1/RELEASE_PPC Power Macintosh
	  % gcc --version
	  powerpc-apple-darwin9-gcc-4.0.1 (GCC) 4.0.1 (Apple Inc. build 5493)
	  Copyright (C) 2005 Free Software Foundation, Inc.
	  This is free software; see the source for copying conditions.  There is NO
	  warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

	building in place enters into an infinite recursion on "#include_next":

	  % gmake V=1
	  [snip]
	  depbase=`echo lib/math.o | sed 's|[^/]*$|.deps/&|;s|\.o$||'`;\
	  gcc -std=gnu99    -I. -Ilib -I. -I./lib   -g -O2 -MT lib/math.o -MD -MP -MF $depbase.Tpo -c -o lib/math.o lib/math.c &&\
	  mv -f $depbase.Tpo $depbase.Po
	  In file included from lib/math.h:27,
	                   from lib/math.h:27,
	                   from lib/math.h:27,
	                   from lib/math.h:27,
	  [snip]
	                   from lib/math.h:27,
	                   from lib/math.h:27,
	                   from lib/math.c:3:
	  lib/math.h:27:23: error: #include nested too deeply
	  Makefile:3414: recipe for target 'lib/math.o' failed
	  gmake[2]: *** [lib/math.o] Error 1

	Using -I./lib instead of -Ilib fixes the problem.

	Reported by Pho.
	<https://lists.gnu.org/archive/html/bison-patches/2014-01/msg00000.html>

	* Makefile.am (AM_CPPFLAGS): Use -I./lib instead of -Ilib.

2015-01-16  Akim Demaille  <akim@lrde.epita.fr>

	doc: minor fixes
	* doc/bison.texi: Fix warnings about colon in reference names.
	* data/bison.m4, src/files.h: Fix comments.
	* doc/Doxyfile.in: update.

2015-01-15  Akim Demaille  <akim@lrde.epita.fr>

	gnulib: strtoul is considered obsolete and now useless
	* bootstrap.conf: here.

2015-01-15  Akim Demaille  <akim@lrde.epita.fr>

	c++: avoid warnings when destructors don't use $$
	* data/c++.m4: here.

2015-01-15  Akim Demaille  <akim@lrde.epita.fr>

	maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2015-01-15  Akim Demaille  <akim@lrde.epita.fr>

	version 3.0.3
	* NEWS: Record release date.

2015-01-15  Akim Demaille  <akim@lrde.epita.fr>

	gnulib: update

2015-01-13  Akim Demaille  <akim@lrde.epita.fr>

	tests: split a large test case into several smaller ones
	* tests/conflicts.at (AT_CONSISTENT_ERRORS_CHECK): Move AT_SETUP/AT_CLEANUP
	into it, so that we don't skip non Java tests following a test case in Java.

2015-01-12  Akim Demaille  <akim@lrde.epita.fr>

	package: a bit of trouble shooting indications
	* README-hacking: here.

2015-01-12  Akim Demaille  <akim@lrde.epita.fr>

	doc: liby's main arms the internationalization
	Reported by Nicolas Bedon.
	<https://lists.gnu.org/archive/html/bug-bison/2014-11/msg00005.html>

	* doc/bison.texi (Yacc Library): Document the call the setlocale.

2015-01-09  Akim Demaille  <akim@lrde.epita.fr>

	bison: avoid warnings from static code analysis
	A static analysis tool reports that some callers of symbol_list_n_get
	might get NULL and not handle it properly.  This is not the case, yet
	we can suppress this pattern.

	Reported by Mike Sullivan.
	<https://lists.gnu.org/archive/html/bug-bison/2013-12/msg00027.html>

	* src/symlist.c (symbol_list_n_get): Actually it is never called
	to return 0.  Enforce this postcondition via aver.
	(symbol_list_n_type_name_get): Simplify accordingly.  In particular,
	discards a (translated) useless error message.
	* src/symlist.h: Adjust documentation.
	* src/scan-code.l: Style change.

2015-01-09  Akim Demaille  <akim@lrde.epita.fr>

	c++: fix the use of destructors when variants are enabled
	When using variants, destructors generate invalid code.
	<http://lists.gnu.org/archive/html/bug-bison/2014-09/msg00005.html>
	Reported by Michael Catanzaro.

	* data/c++.m4 (~basic_symbol): b4_symbol_foreach works on yysym:
	define it.
	* tests/c++.at (Variants): Check it.

2015-01-08  Akim Demaille  <akim@lrde.epita.fr>

	style: tests: simplify the handling of some C++ tests
	* tests/c++.at: here.
	(Doxygen): Pass %define, so that files such as position.hh etc.
	are generated, instead of putting everything into input.hh.

2015-01-08  Akim Demaille  <akim@lrde.epita.fr>

	c++: symbols can be empty, so use it
	The previous patches ensure that symbols (symbol_type and
	stack_symbol_type) can be empty, cleared, and their emptiness can be
	checked.  Therefore, yyempty, which codes whether yyla is empty or
	not, is now useless.

	In C skeletons (e.g., yacc.c), the fact that the lookahead is empty is
	coded by "yychar = YYEMPTY", which is exactly what this patch
	restores, since yychar/yytoken corresponds to yyla.type.

	* data/lalr1.cc (yyempty): Remove.
	Rather, depend on yyla.empty ().

2015-01-08  Akim Demaille  <akim@lrde.epita.fr>

	c++: variants: don't leak the lookahead in error recovery
	During error recovery, when discarding the lookeahead, we don't
	destroy it, which is caught by parse.assert assertions.

	Reported by Antonio Silva Correia.
	With an analysis and suggested patch from Michel d'Hooge.
	<http://savannah.gnu.org/support/?108481>

	* tests/c++.at (Variants): Strengthen the test to try syntax errors
	with discarded lookahead.

2015-01-08  Akim Demaille  <akim@lrde.epita.fr>

	c++: provide a means to clear symbols
	The symbol destructor is currently the only means to clear a symbol.
	Unfortunately during error recovery we might have to clear the
	lookahead, which is a local variable (yyla) that has not yet reached
	its end of scope.

	Rather that duplicating the code to destroy a symbol, or rather than
	destroying and recreating yyla, let's provide a means to clear a
	symbol.

	Reported by Antonio Silva Correia, with an analysis from Michel d'Hooge.
	<http://savannah.gnu.org/support/?108481>

	* data/c++.m4, data/lalr1.cc (basis_symbol::clear, by_state::clear)
	(by_type::clear): New.
	(basic_symbol::~basic_symbol): Use clear.

2015-01-08  Akim Demaille  <akim@lrde.epita.fr>

	c++: clean up the handling of empty symbols
	* data/c++.m4, data/lalr1.cc (yyempty_): Remove, replaced by...
	(empty_symbol, by_state::empty_state): these.
	(basic_symbol::empty): New.

2015-01-08  Akim Demaille  <akim@lrde.epita.fr>

	c++: comment and style changes
	* data/c++.m4, data/lalr1.cc: More documentation.
	Tidy.
	* tests/c++.at (string_cast): Rename as...
	(to_string): this C++11 name.

2015-01-07  Akim Demaille  <akim@lrde.epita.fr>

	c++: variants: comparing addresses of typeid.name() is undefined
	Instead of storing and comparing pointers to names of types, store
	pointers to the typeids, and compares the typeids.
	Reported by Thomas Jahns.
	<http://lists.gnu.org/archive/html/bug-bison/2014-03/msg00001.html>

	* data/variant.hh (yytname_): Replace with...
	(yytypeid_): this.

2015-01-05  Akim Demaille  <akim@lrde.epita.fr>

	c++: locations: complete the API and fix comments
	There are no support for += between locations, and some comments are wrong.
	Reported by Alexandre Duret-Lutz.

	* data/location.cc: Fix.
	* doc/bison.texi: Document.
	* tests/c++.at: Check.

2015-01-05  Akim Demaille  <akim@lrde.epita.fr>

	build: do not clean figure sources in make clean
	"make clean && make" fails in in-tree builds.

	* doc/local.mk (CLEANDIRS): Replace with...
	(CLEANFILES): this safer list of files to clean.

2015-01-05  Akim Demaille  <akim@lrde.epita.fr>

	build: don't try to generate docs when cross-compiling
	When cross-compiling don't run the generated bison to update the docs.
	Reported by Aaro Koskinen.
	<http://lists.gnu.org/archive/html/bison-patches/2014-03/msg00000.html>

	* configure.ac (CROSS_COMPILING): New.
	* doc/local.mk: Use it.

2015-01-04  Akim Demaille  <akim@lrde.epita.fr>

	package: fix a reporter's name
	* THANKS, build-aux/git-log-fix: s/Bernd Edligner/Bernd Edlinger/.

2015-01-04  Akim Demaille  <akim@lrde.epita.fr>

	%union: fix the support for named %union
	Bison supports a union tag, for obscure reasons.  But it does a poor
	job at it, especially since Bison 3.0.
	Reported by Stephen Cameron and Tobias Frost.

	It did not ensure that the name was not given several times.  An easy
	way to do this is to make the %union tag be handled as a %define
	variable, as they cannot be defined several times.

	Since Bison 3.0, the synclines were wrongly placed, resulting in
	invalid code.  Addressing this issue, because of the way the union tag
	was stored (as a code muscle), would have been tedious.  Unless we
	rather define the %union tag as a %percent variable, whose synclines
	are easier to manipulate.

	So replace the b4_union_name muscle by the api.value.union.name
	%define variable, document, and check.

	* data/bison.m4: Make sure that api.value.union.name has a keyword value.
	* data/c++.m4: Make sure that api.value.union.name is not defined.
	* data/c.m4 (b4_union_name): No longer use it, use api.value.union.name.
	* doc/bison.texi (%define Summary): Document it.
	* src/parse-gram.y (union_name): No longer define b4_uion_name, but
	api.value.union.name.
	* tests/input.at (Redefined %union name): New.
	* tests/synclines.at (%union name syncline): New.
	* tests/types.at: Check named %unions.

2015-01-04  Akim Demaille  <akim@lrde.epita.fr>

	package: bump to 2015
	Which also requires:

	* gnulib: Update.

2014-12-31  Akim Demaille  <akim@lrde.epita.fr>

	flex: don't trust YY_USER_INIT
	Reported by Bernd Edlinger and others.

	* src/scan-gram.l: here.

2014-12-31  Akim Demaille  <akim@lrde.epita.fr>

	yacc.c: fix broken union when api.value.type=union and %defines are used
	Reported by Rich Wilson.

	* data/c.m4 (b4_symbol_type_register): Append to b4_union_members,
	not b4_user_union_members.
	The latter invokes the former, but it is the former which is reinitialized
	to empty by b4_value_type_setup_union.
	* tests/types.at: Check it.

	This reveals another bug, this time in the case of glr.c parsers.

	* data/glr.c: Generate the header file before the implementation file,
	to be sure that the setup is run before what depends on it.

2014-12-31  Akim Demaille  <akim@lrde.epita.fr>

	doc: fix missing xref
	Reported by xolodho.

	* doc/bison.texi (Printer Decl): here.

2014-12-29  Akim Demaille  <akim@lrde.epita.fr>

	gnulib: update

2014-02-03  Akim Demaille  <akim@lrde.epita.fr>

	location: remove some ugly debugging code traces
	* data/location.cc: here.

2014-02-03  Akim Demaille  <akim@lrde.epita.fr>

	build: use abort to pacify compiler errors
	clang, with -DNDEBUG and -Werror fails on some functions that might
	lack a return.  This is because aver is just another assert, discarded
	with -DNDEBUG.  So use abort.

	* src/muscle-tab.c, src/scan-skel.l: here.

2014-02-03  Akim Demaille  <akim@lrde.epita.fr>

	package: bump to 2014
	* AUTHORS, ChangeLog-2012, Makefile.am, NEWS, PACKAGING, README,
	* README-alpha, README-hacking, THANKS, TODO, bootstrap.conf,
	* build-aux/darwin11.4.0.valgrind, build-aux/local.mk,
	* build-aux/update-b4-copyright,
	* build-aux/update-package-copyright-year, cfg.mk, configure.ac,
	* data/README, data/bison.m4, data/c++-skel.m4, data/c++.m4,
	* data/c-like.m4, data/c-skel.m4, data/c.m4, data/glr.c, data/glr.cc,
	* data/java-skel.m4, data/java.m4, data/lalr1.cc, data/lalr1.java,
	* data/local.mk, data/location.cc, data/stack.hh, data/variant.hh,
	* data/xslt/bison.xsl, data/xslt/xml2dot.xsl, data/xslt/xml2text.xsl,
	* data/xslt/xml2xhtml.xsl, data/yacc.c, djgpp/Makefile.maint,
	* djgpp/README.in, djgpp/config.bat, djgpp/config.sed,
	* djgpp/config.site, djgpp/config_h.sed, djgpp/djunpack.bat,
	* djgpp/local.mk, djgpp/subpipe.c, djgpp/subpipe.h,
	* djgpp/testsuite.sed, doc/bison.texi, doc/local.mk, doc/refcard.tex,
	* etc/README, etc/bench.pl.in, etc/local.mk,
	* examples/calc++/calc++.test, examples/calc++/local.mk,
	* examples/extexi, examples/local.mk, examples/mfcalc/local.mk,
	* examples/mfcalc/mfcalc.test, examples/rpcalc/local.mk,
	* examples/rpcalc/rpcalc.test, examples/test, examples/variant.yy,
	* lib/abitset.c, lib/abitset.h, lib/bbitset.h, lib/bitset.c,
	* lib/bitset.h, lib/bitset_stats.c, lib/bitset_stats.h,
	* lib/bitsetv-print.c, lib/bitsetv-print.h, lib/bitsetv.c,
	* lib/bitsetv.h, lib/ebitset.c, lib/ebitset.h, lib/get-errno.c,
	* lib/get-errno.h, lib/lbitset.c, lib/lbitset.h, lib/libiberty.h,
	* lib/local.mk, lib/main.c, lib/timevar.c, lib/timevar.def,
	* lib/timevar.h, lib/vbitset.c, lib/vbitset.h, lib/yyerror.c,
	* m4/bison-i18n.m4, m4/c-working.m4, m4/cxx.m4, m4/flex.m4,
	* m4/timevar.m4, src/AnnotationList.c, src/AnnotationList.h,
	* src/InadequacyList.c, src/InadequacyList.h, src/LR0.c, src/LR0.h,
	* src/Sbitset.c, src/Sbitset.h, src/assoc.c, src/assoc.h,
	* src/closure.c, src/closure.h, src/complain.c, src/complain.h,
	* src/conflicts.c, src/conflicts.h, src/derives.c, src/derives.h,
	* src/files.c, src/files.h, src/flex-scanner.h, src/getargs.c,
	* src/getargs.h, src/gram.c, src/gram.h, src/graphviz.c,
	* src/graphviz.h, src/ielr.c, src/ielr.h, src/lalr.c, src/lalr.h,
	* src/local.mk, src/location.c, src/location.h, src/main.c,
	* src/muscle-tab.c, src/muscle-tab.h, src/named-ref.c,
	* src/named-ref.h, src/nullable.c, src/nullable.h, src/output.c,
	* src/output.h, src/parse-gram.c, src/parse-gram.y, src/print-xml.c,
	* src/print-xml.h, src/print.c, src/print.h, src/print_graph.c,
	* src/print_graph.h, src/reader.c, src/reader.h, src/reduce.c,
	* src/reduce.h, src/relation.c, src/relation.h, src/scan-code.h,
	* src/scan-code.l, src/scan-gram.h, src/scan-gram.l, src/scan-skel.h,
	* src/scan-skel.l, src/state.c, src/state.h, src/symlist.c,
	* src/symlist.h, src/symtab.c, src/symtab.h, src/system.h,
	* src/tables.c, src/tables.h, src/uniqstr.c, src/uniqstr.h,
	* tests/actions.at, tests/atlocal.in, tests/bison.in, tests/c++.at,
	* tests/calc.at, tests/conflicts.at, tests/cxx-type.at,
	* tests/existing.at, tests/glr-regression.at, tests/headers.at,
	* tests/input.at, tests/java.at, tests/javapush.at, tests/local.at,
	* tests/local.mk, tests/named-refs.at, tests/output.at, tests/push.at,
	* tests/reduce.at, tests/regression.at, tests/sets.at,
	* tests/skeletons.at, tests/synclines.at, tests/testsuite.at,
	* tests/torture.at, tests/types.at:
	here.

2014-01-03  Paul Eggert  <eggert@cs.ucla.edu>

	doc: specify documentation encoding
	* doc/bison.texi: Add '@documentencoding UTF-8'; needed since the
	manual contains UTF-8 characters.  This will cause the .info files
	to contain UTF-8 quotes and the like, which should be OK nowadays.
	Add @documentlanguage while we're at it.

2013-12-09  Akim Demaille  <akim@lrde.epita.fr>

	package: install the examples
	Currently, we do not install the various examples extracted from the
	documentation.  Let's do it, as they are useful starting points.

	* configure.ac: When --enable-gcc-warnings is set, enable ENABLE_GCC_WARNINGS.
	* examples/extexi: No longer issue synclines by default.
	* examples/local.mk: Except if ENABLE_GCC_WARNINGS.
	* examples/calc++/local.mk, examples/mfcalc/local.mk,
	* examples/rpcalc/local.mk: Install the example files.

2013-12-09  Akim Demaille  <akim@lrde.epita.fr>

	package: install README and the like in docdir
	* Makefile.am: here.

2013-12-09  Akim Demaille  <akim@lrde.epita.fr>

	diagnostics: fix the order of multiple declarations reports
	On

	  %token FOO "foo"
	  %printer {} "foo"
	  %printer {} FOO

	we report

	  /tmp/foo.yy:2.10-11: error: %printer redeclaration for FOO
	   %printer {} "foo"
	            ^^
	  /tmp/foo.yy:3.10-11:     previous declaration
	   %printer {} FOO
	            ^^

	* src/symtab.c (locations_sort): New.
	Use it.
	* tests/input.at (Invalid Aliases): Stress the order of diagnostics.

2013-12-09  Akim Demaille  <akim@lrde.epita.fr>

	symbol: provide an easy means to compare them in source order
	* src/symtab.c (symbols_sort): New.
	(user_token_number_redeclaration): Taken from here.

2013-12-05  Akim Demaille  <akim@lrde.epita.fr>

	maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2013-12-05  Akim Demaille  <akim@lrde.epita.fr>

	version 3.0.2
	* NEWS: Record release date.

2013-12-05  Akim Demaille  <akim@lrde.epita.fr>

	gnulib: update
	* gnulib: here.

2013-12-04  Akim Demaille  <akim@lrde.epita.fr>

	output: do not generate source files when late errors are caught
	Reported by Alexandre Duret-Lutz as "second problem" in:
	http://lists.gnu.org/archive/html/bug-bison/2013-09/msg00015.html

	* bootstrap.conf: We need the "unlink" module.
	* src/files.h, src/files.c (unlink_generated_sources): New.
	* src/output.c: Use it.
	* tests/output.at: Check the case of late errors.

2013-12-04  Akim Demaille  <akim@lrde.epita.fr>

	output: record what generated files are source or report files
	* src/files.h, src/files.c (output_file_name_check): Take an additional
	argument to record whether a file is a source or report file.
	* src/files.c (generated_file): New.
	(file_names, file_names_count): Replace with...
	(generated_files, generated_files_size): these.
	* src/scan-skel.l: Adjust.

2013-12-04  Akim Demaille  <akim@lrde.epita.fr>

	output: do not generate source files when early errors are caught
	Reported by Alexandre Duret-Lutz as "second problem" in:
	http://lists.gnu.org/archive/html/bug-bison/2013-09/msg00015.html

	One problem is that some errors are caught early, before the
	generation of output files, while others can only be detected
	afterwards (since, for instance, skeletons can raise errors
	themselves).

	This will be addressed in two steps: early errors do not generate
	source files at all, while later errors will remove the files that
	have already been generated.

	* src/scan-skel.l (yyout): Open to /dev/null when there are errors.
	* tests/output.at (AT_CHECK_FILES): Factored out of...
	(AT_CHECK_OUTPUT): this.
	Fuse the "SHELLIO" argument in the "FLAGS" one.
	Use $5 to denote the expected exit status.
	Add a test case for early errors.

2013-11-26  Akim Demaille  <akim@lrde.epita.fr>

	xml: also use "%empty" with html output
	* data/xslt/xml2xhtml.xsl: No longer issue an Epsilon, display as in
	dot and text formats.

2013-11-26  Akim Demaille  <akim@lrde.epita.fr>

	style: formatting changes
	* src/print-xml.c: here.

2013-11-26  Akim Demaille  <akim@lrde.epita.fr>

	xml: also display %empty for empty right-hand sides
	* data/xslt/xml2dot.xsl, data/xslt/xml2text.xsl: Display %empty where needed.

2013-11-26  Akim Demaille  <akim@lrde.epita.fr>

	reports: display %empty in the generated pointed-rules
	* src/print.c (print_core): Use %empty for empty rules.
	* src/print_graph.c (print_core): Ditto.
	* tests/conflicts.at, tests/output.at, tests/reduce.at: Adjust
	expectations.

2013-11-26  Akim Demaille  <akim@lrde.epita.fr>

	news: YYERROR vs variants

2013-11-18  Akim Demaille  <akim@lrde.epita.fr>

	style: scope reduction in lalr.cc
	* src/lalr.c: Shorten variable scopes.
	(lookahead_tokens_print): Use the same variable name in two loops
	iterating over the same structure.

2013-11-15  Akim Demaille  <akim@lrde.epita.fr>

	lalr1.cc: formatting changes
	* data/lalr1.cc: Fix indentation.

2013-11-15  Akim Demaille  <akim@lrde.epita.fr>

	lalr1.cc: fix the support of YYERROR with variants
	When variant are enabled, the yylhs variable (the left-hand side of
	the rule being reduced, i.e. $$ and @$) is explicitly destroyed when
	YYERROR is called.  This is because before running the user code, $$
	is initialized, so that the user can properly use it.

	However, when quitting yyparse, yylhs is also reclaimed by the C++
	compiler: the variable goes out of scope.

	Instead of trying to be too smart, let the compiler do its job: reduce
	the scope of yylhs to exactly the reduction.  This way, whatever the
	type of scope exit (regular, exception, return, goto...) this variable
	will be properly reclaimed.

	Reported by Paolo Simone Gasparello.
	<http://lists.gnu.org/archive/html/bug-bison/2013-10/msg00003.html>

	* data/lalr1.cc (yyparse): Reduce the scope of yylhs.
	* tests/c++.at: We now pass this test.

2013-11-15  Akim Demaille  <akim@lrde.epita.fr>

	tests: check $$'s destruction with variant, YYERROR, and no error recovery
	When variant are enabled, the yylhs variable (the left-hand side of
	the rule being reduced, i.e. $$ and @$) is explicitly destroyed when
	YYERROR is called.  This is because before running the user code, $$
	is initialized, so that the user can properly use it.

	However, when quitting yyparse, yylhs is also reclaimed by the C++
	compiler: the variable goes out of scope.

	This was not detected by the test suite because (i) the Object tracker
	was too weak, and (ii) the problem does not show when there is error
	recovery.

	Reported by Paolo Simone Gasparello.
	<http://lists.gnu.org/archive/html/bug-bison/2013-10/msg00003.html>

	* tests/c++.at (Exception safety): Improve the objects logger to make
	sure that we never destroy twice an object.
	Also track copy-constructors.
	Use a set instead of a list.
	Display the logs before running the function body, this is more
	useful in case of failure.
	Generalize to track with and without error recovery.

2013-11-15  Akim Demaille  <akim@lrde.epita.fr>

	tests: simplify useless obfuscation
	* tests/c++.at: $$ is not special for M4, there is no need to "escape"
	it.

2013-11-14  Akim Demaille  <akim@lrde.epita.fr>

	skeletons: use better names when computing a "goto"
	* data/glr.c (yyLRgotoState): Name the symbol argument yysym, instead
	of yylhs.
	* data/lalr1.cc (yy_lr_goto_state_): Likewise.
	* data/lalr1.java (yy_lr_goto_state_): New, modeled after the previous
	two routines.
	Use it.

2013-11-12  Akim Demaille  <akim@lrde.epita.fr>

	maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2013-11-12  Akim Demaille  <akim@lrde.epita.fr>

	version 3.0.1
	* NEWS: Record release date.

2013-11-12  Akim Demaille  <akim@lrde.epita.fr>

	aver: it is no longer "protected against NDEBUG"
	Apply the same rules for aver as for assert: no side effects,
	especially not important ones.

	* src/AnnotationList.c, src/muscle-tab.c: Adjust aver uses to resist
	to -DNDEBUG.

2013-11-08  Akim Demaille  <akim@lrde.epita.fr>

	parsers: rename YY_NULL as YY_NULLPTR to avoid conflicts with Flex
	Flex also defines YY_NULL (to 0).  Avoid gratuitous conflicts.

	* data/c.m4 (b4_null_define): Rename YY_NULL as YY_NULLPTR.

	* data/glr.c, data/lalr1.cc, data/location.cc, data/variant.hh,
	* data/yacc.c, src/parse-gram.c, tests/actions.at, tests/c++.at,
	* tests/cxx-type.at, tests/glr-regression.at, tests/headers.at,
	* tests/push.at, tests/regression.at:
	Adjust.

2013-11-05  Akim Demaille  <akim@lrde.epita.fr>

	build: use Automake 1.14's non-recursive Makefile features
	* configure.ac: Require Automake 1.14.
	* examples/calc++/local.mk, examples/local.mk, examples/mfcalc/local.mk,
	* examples/rpcalc/local.mk, tests/local.mk: Use %D% and %C%.

2013-11-05  Akim Demaille  <akim@lrde.epita.fr>

	build: restore maintainer-push-check
	* tests/local.mk: here.

2013-11-05  Akim Demaille  <akim@lrde.epita.fr>

	c++: use __attribute__((__pure__)) to avoid warnings
	Building C++ parsers with -Wsuggest-attribute=const and
	-Wsuggest-attribute=noreturn triggers warning in generated code.

	* data/lalr1.cc: Call b4_attribute_define.
	(debug_stream, debug_level): Flag as pure.
	* tests/headers.at (Several parsers): There are now more YY macros
	that "leak".

2013-11-05  Akim Demaille  <akim@lrde.epita.fr>

	skeletons: update the handling of compiler attributes
	* data/c.m4 (b4_attribute_define): Instead of defining __attribute__,
	define YY_ATTRIBUTE conditionally.
	(YY_ATTRIBUTE_PURE, YY_ATTRIBUTE_UNUSED, _Noreturn): New.
	Use them.
	* data/glr.c: Use them.

2013-11-05  Akim Demaille  <akim@lrde.epita.fr>

	gnulib: update

2013-10-24  Akim Demaille  <akim@lrde.epita.fr>

	style: use /* ... */ comments
	* src/complain.c: Here.

2013-10-24  Akim Demaille  <akim@lrde.epita.fr>

	tests: skip C++ tests that are too demanding for some compilers
	Some tests now fail when compiled with G++ 4.3 or 4.4 on MacPorts.

	* tests/local.at (AT_SKIP_IF_EXCEPTION_SUPPORT_IS_POOR): New.
	* tests/c++.at (Exception safety): Use it.

2013-10-22  Akim Demaille  <akim@lrde.epita.fr>

	install: do not install yacc.1 when --disable-yacc
	* configure.ac (ENABLE_YACC): New conditional.
	(YACC_SCRIPT, YACC_LIBRARY): Remove.
	* lib/local.mk, src/local.mk: Use the former instead of the latter.
	* doc/local.mk: Use ENABLE_YACC to avoid installing yacc.1.

2013-10-22  Akim Demaille  <akim@lrde.epita.fr>

	style: avoid tabs
	* src/scan-code.l: here.

2013-10-22  Akim Demaille  <akim@lrde.epita.fr>

	c++: fix generated doxygen comments
	* configure.ac: Enable -Wdocumentation if supported.
	* data/lalr1.cc: Fix comments.

2013-10-22  Akim Demaille  <akim@lrde.epita.fr>

	fix: uniqstr are already pointers
	* src/uniqstr.c (uniqstr_assert): Remove incorrect double indirection,
	and now useless cast.

2013-10-22  Paul Eggert  <eggert@cs.ucla.edu>

	bison: pacify Sun C 5.12
	* src/scan-code.l (show_sub_message):
	Redo initializations to work around a bogus Sun C 5.12 warning.
	(parse_ref): Remove unreachable code that Sun C 5.12 complains about.
	* src/uniqstr.h (uniqstr_vsprintf): Use
	_GL_ATTRIBUTE_FORMAT_PRINTF (...)  instead of __attribute__
	((__format__ (__printf__, ...))).  Otherwise, Sun C 5.12
	complains about an unknown attribute.

2013-10-22  Paul Eggert  <eggert@cs.ucla.edu>

	maint: git now ignores rpcalc
	* examples/rpcalc/.gitignore: Ignore rpcalc.

2013-10-22  Paul Eggert  <eggert@cs.ucla.edu>

	build: examples/calc++/calc++ requires flex
	* configure.ac (FLEX_CXX_WORKS): New AM_CONDITIONAL.
	* examples/calc++/local.mk (examples/calc++/calc++):
	Build if FLEX_CXX_WORKS, not BISON_CXX_WORKS.

2013-10-22  Paul Eggert  <eggert@cs.ucla.edu>

	maint: mention help2man, texinfo, apt-get
	* README-hacking: Add help2man, texinfo.
	Describe how to add packages if you're using Debian.

2013-10-22  Paul Eggert  <eggert@cs.ucla.edu>

	maint: git now ignores .log and .trs files
	* .gitignore: Add *.log, *.trs.

2013-10-21  Akim Demaille  <akim@lrde.epita.fr>

	tests: fix incorrect object construction
	Reported by Ken Moffat.
	http://lists.gnu.org/archive/html/bug-bison/2013-10/msg00009.html

	* tests/c++.at (Exception safety): Here.

2013-10-16  Akim Demaille  <akim@lrde.epita.fr>

	glr: allow spaces between "%?" and "{" in predicates
	Reported by Rici Lake.
	http://lists.gnu.org/archive/html/bug-bison/2013-10/msg00004.html
	http://stackoverflow.com/questions/19330171/

	* src/scan-gram.l: Do not try to be too smart when diagnosing invalid
	directives.
	* tests/glr-regression.at (Predicates): New test.

2013-10-16  Akim Demaille  <akim@lrde.epita.fr>

	diagnostics: "-Werror -Wno-error=foo" must not emit errors
	Currently "-Werror -Wno-error=foo" still turns "foo" warnings into errors.
	Reported by Alexandre Duret-Lutz.
	See http://lists.gnu.org/archive/html/bug-bison/2013-09/msg00015.html.

	* src/complain.c (errority, errority_flag): New.
	(complain_init): Initialize the latter.
	(warning_argmatch): Extract the loop iterating on the flag's bits.
	Set and unset errority_flag here.
	(warnings_argmatch): -Wno-error is not the same as -Wno-error=everything:
	we must remember if category foo was explicitly turned in an error/warning
	via -W(no-)error=foo.
	(warning_severity): Use errority_flag.

	* tests/input.at (Symbols): Just check --yacc, not -Wyacc, that's the
	job of tests on -W.
	(-Werror is not affected by -Wnone and -Wall): Rename as...
	(-Werror combinations): this.
	Tests more combinations of -W, -W(no-)error, and -W(no-)error=foo.
	* tests/local.at (AT_BISON_CHECK_WARNINGS): Don't expect -Werror
	to turn runs that issue warnings into runs with errors, as the
	warnings might be enforced as warnings by -Wno-error=foo, in which
	case -Werror does not change anything.

	* doc/bison.texi (Bison Options): Try to be clearer about how
	-W(no-)error and -W(no-)error=foo interact.

2013-10-16  Akim Demaille  <akim@lrde.epita.fr>

	comment changes
	* src/complain.h, src/complain.c: More documentation, more comments.

2013-10-04  Andreas Schwab  <schwab@linux-m68k.org>

	location: fix EOF check
	* location.c (location_caret): Use int, not char, for values from
	getc.

2013-09-19  Akim Demaille  <akim@lrde.epita.fr>

	style: variant: remove empty line
	* data/variant.hh (b4_symbol_constructor_define_): Remove
	stray eol.

2013-09-19  Akim Demaille  <akim@lrde.epita.fr>

	glr: more assertions
	* data/glr.c (yyaddDeferredAction, yyglrShiftDefer, yypdumpstack):
	More assertions.

2013-09-19  Akim Demaille  <akim@lrde.epita.fr>

	glr: shorten scopes
	* data/glr.c (yyglrReduce): Define yyflag with its value.

2013-09-19  Akim Demaille  <akim@lrde.epita.fr>

	glr: formatting changes
	* data/glr.c: here.

2013-09-19  Akim Demaille  <akim@lrde.epita.fr>

	glr: better use of tracing macros
	* data/glr.c (yydestroyGLRState): Use YY_SYMBOL_PRINT instead of
	yy_symbol_print.

2013-09-19  Akim Demaille  <akim@lrde.epita.fr>

	examples: improve the output of the "variant" example
	* examples/variant.yy: Improve the printing of lists.

2013-09-19  Akim Demaille  <akim@lrde.epita.fr>

	variant: remove useless assertion
	* data/variant.hh (move): Remove precondition assertion which is
	ensured by the first call of the body (this precondition is also one
	of "build").

2013-09-19  Akim Demaille  <akim@lrde.epita.fr>

	tests: remove stray debugging traces
	* tests/atlocal.in: Remove traces.
	Be ready to remove conftest.dSYM generated on OS X.

2013-09-04  Akim Demaille  <akim@lrde.epita.fr>

	tests: do not use grep -q
	Reported by Daniel Galloway.
	http://lists.gnu.org/archive/html/bug-bison/2013-08/msg00020.html

	* tests/java.at: Ignore grep's output instead.

2013-08-25  Paul Eggert  <eggert@cs.ucla.edu>

	build: don't require flex for ordinary builds
	* configure.ac (LEX): Don't fail if this is lex, as flex is not
	required for ordinary builds.  Instead, issue a warning and
	substitute a no-op LEX.  Reported by Michael Felt in
	<http://lists.gnu.org/archive/html/bug-bison/2013-08/msg00009.html>.

2013-08-25  Paul Eggert  <eggert@cs.ucla.edu>

	maint: update .gitignore
	* .gitignore: Add *.eps, *.o, *.pdf, *.png, *.stamp, *~,
	.deps, .dirstamp.  Needed to suppress unwanted chatter from
	'git status' after a bootstrap build.

2013-08-24  Paul Eggert  <eggert@cs.ucla.edu>

	build: port to pre-5.8.7 perl
	* examples/local.mk (extract): Omit -f from perl options.
	This doesn't work with perl versions before 5.8.7
	that are configured without USE_SITECUSTOMIZE.
	Reported by Michael Felt in
	<http://lists.gnu.org/archive/html/bug-bison/2013-08/msg00006.html>.

2013-08-01  Akim Demaille  <akim@lrde.epita.fr>

	tests: minor change to make it easier to test other skeletons
	* tests/c++.at (Variants): Pass the skeleton as argument.

2013-08-01  Valentin Tolmer  <valentin.tolmer@gmail.com>

	uniqstr: fix assertion
	* src/uniqstr.c (uniqstr_assert): Really make sure str is a uniqstr,
	not just whether some uniqstr with the same content was registered.
